Refactor indices pallet to use fungible traits #9610
Conversation
…s://github.com/Krayt78/polkadot-sdk into Refactor-Indices-pallet-to-use-fungible-traits
|
Asking for reviews to run the bot so i can correct whats missing and also having some insight on if what i am doing is correct or not as this is the first time i am doing a migration so i am pretty sure i am doing something wrong |
…ght handling during account migration steps. Updated function names for clarity and adjusted weight consumption checks.
|
/cmd label T2-pallets |
paritytech-cmd-bot-polkadot-sdk
bot
added
the
T2-pallets
| return Err(SteppedMigrationError::InsufficientWeight { required: min_required }); | ||
| } | ||
|
|
||
| loop { |
There was a problem hiding this comment.
I wonder if it is standard to do a loop inside a step, we could also just do one step which is migration one account or a maximum number of account.
Probably both ways are fine.
There was a problem hiding this comment.
I saw other migrations doing loops in the step. i am not sure on how these migrations work, is it one step per block ? Cause if it is it would make the migration quite slow to do them 1 by 1.
I could just look at the remaining weight available and calculate how many migrations i can do in that step and remove the loop
There was a problem hiding this comment.
I would expect they do steps until the block is filled.
There was a problem hiding this comment.
actually I don't exactly know how multi step works. but if I was designing it I would have made as many steps are possible in a block, in a way programmable by the block author so that is the step weight is underestimating then it doesn't get stuck but instead the block author can force to decrease the number of steps.
What I am scared about is that we loop to fill the weight but actually our benchmark underestimate the weight and the actual weight is too big for the parachain block, and then the block production is stuck.
There was a problem hiding this comment.
after some digging it seems that the framework calls step() once per block.
so to me using a loop iinside the step to empty all the weight available is the correct pattern
…migration benchmark to avoid benchmarking it everytime
…rate for account storage insertion.
2 participants
Overview
This PR migrates the indices pallet from the Currency trait to the fungible traits with holds, implementing a comprehensive migration strategy that preserves all index relationships and ensures no funds are lost during the transition.
Changes
Implemented MigrateCurrencyToFungibles migration that converts from Currency reserves to Fungibles holds
Migration uses stepped execution to handle large numbers of accounts without timing out
Added detailed logging throughout the migration process for debugging and monitoring
Storage Changes
Updated to use the existing Accounts storage with Fungibles holds
Test Coverage
Updated the new tests to be compatible with the changes
Resources
This is part of the umbrella task #226.
Migration
I have setup a draft migration case but still need to test with some fork of the chain to see how it goes with real life data.